TWI693267B - Adhesive and connecting structure - Google Patents

Abstract

The present invention provides an adhesive having excellent adhesion and heat dissipation to an oxide film, and a connection structure using the same.

The adhesive contains an epoxy compound, a cationic catalyst, and an acrylic resin containing acrylic acid and an acrylate having a hydroxyl group. The acrylic acid in the acrylic resin reacts with the epoxy compound to produce a connection between the island 13 of the acrylic resin and the sea 12 of the epoxy compound, and roughens the surface of the oxide film 11a to enhance the anchoring effect with the sea 12 of the epoxy compound, and The solder particles 1 contained therein are melted, thereby forming a metal bond with the electrode 10, improving the adhesion between the adhesive and the electrode 10, and further improving the heat dissipation characteristics from the metal bonding surface.

Description

Adhesive and connecting structure

The invention relates to an adhesive that electrically connects electronic components to each other, in particular to an adhesive that connects a heat-generating electronic component to a wiring board and dissipates heat of the electronic component, and an electronic component is connected to the wiring board Connect the structure.

This application is based on the Japanese Patent Application No. Japanese Patent Application No. 2014-107167 filed in Japan on May 23, 2014 and the Japanese Patent Application No. Japanese Patent Application No. 2014-107168 filed on May 23, 2014. If the right is right, these applications are cited in this application by reference.

As a method of assembling wafer components such as LEDs on a circuit board, flip-chip mounting using anisotropic conductive film (ACF: Anisotropic Conductive Film) in which conductive particles are dispersed in an epoxy adhesive and formed into a film shape is widely used A method of wafer mounting (for example, refer to Patent Documents 1 and 2). According to this method, the electrical connection between the chip component and the circuit board is achieved by the conductive particles of the anisotropic conductive film, so the connection process can be shortened and the production efficiency can be improved.

[Prior Technical Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Application Publication No. 2010-24301

[Patent Document 2] Japanese Unexamined Patent Publication No. 2012-186322

In recent years, LED products have changed the metal of the wiring of the circuit board from Au, Ag to Al, Cu in order to achieve low cost, and ITO (Indium Tin) is formed on PET (Polyethylene terephthalate) substrate Oxide) Wiring transparent substrate.

However, on the surface of metal wirings such as Al and Cu or ITO wirings, oxides such as a passive state and an oxide film are formed. Therefore, the previous epoxy-based adhesive system is difficult to adhere.

In addition to the difficulty in adhesion, in order to fully dissipate heat from electronic components that generate heat, such as LED products, it is necessary to include a material for heat dissipation in the adhesive. Since the material for heat dissipation is included, the component of the adhesive is reduced, which makes it difficult to maintain the adhesion sufficiently. .

In addition, if the adhesive contains an inorganic filler or a metal filler used as a material for heat dissipation, the inorganic filler or the metal filler becomes a spacer, and the adhesive layer cannot be thinned.

The present invention is to solve the above-mentioned problems of the prior art, and an object of the present invention is to provide an adhesive having excellent adhesion to an oxide film and excellent heat dissipation from an electronic component that radiates heat to the outside, and a connection structure using the same.

In order to solve the above-mentioned problems, the adhesive of the present invention connects a heat-generating electronic component to a substrate having a wiring pattern, and is characterized by being composed of a resin binder containing solder particles.

In addition, the connection structure of the present invention is characterized by including: a substrate having a wiring pattern; an anisotropic conductive film formed on the electrode of the wiring pattern; and a heat-generating electronic component built on the anisotropic conductive film On; and the anisotropic conductive film contains a resin binder and solder particles, the solder particles and the metal part of the terminal part of the electronic component.

In order to solve the above-mentioned problems, the adhesive of the present invention is characterized by containing an alicyclic epoxy compound or a hydrogenated epoxy compound, a cationic catalyst, an acrylic resin having a weight average molecular weight of 50,000 to 900,000, and solder particles, acrylic resin Contains 0.5~10wt% acrylic acid and 0.5~10wt% acrylic acid ester with hydroxyl group.

Moreover, the connection structure of the present invention is characterized by including: a substrate having a wiring pattern whose surface is made of oxide; an anisotropic conductive film formed on the electrode of the above-mentioned wiring pattern; and an electronic component, its structure On the anisotropic conductive film; and the anisotropic conductive film is a cured product of anisotropic conductive adhesive, the anisotropic conductive adhesive contains an alicyclic epoxy compound or hydrogenated epoxy compound, cationic catalyst, weight Acrylic resins, conductive particles, and solder particles having an average molecular weight of 50,000 to 900,000. The acrylic resins include 0.5 to 10% by weight of acrylic acid and 0.5 to 10% by weight of acrylates with hydroxyl groups.

According to the present invention, the solder particles in the resin adhesive bond to the metal of the terminal part of the electronic part, thereby obtaining excellent adhesion between the adhesive layer and the electronic part, and diffusing the heat generated in the electronic part to the metal bond The solder particles can dissipate heat more efficiently. In addition, according to the present invention, by blending an acrylic resin containing acrylic acid and an acrylate having a hydroxyl group, the oxide film can be bonded as a whole with a cured product, and excellent adhesive force can be obtained, and the bonding can be sufficiently ensured by solder particles. strength.

1‧‧‧ solder particles

1a‧‧‧melted solder

1b‧‧‧End face (metal joint face)

2‧‧‧ conductive particles

3‧‧‧Resin adhesive

10‧‧‧electrode

11‧‧‧Wiring

11a‧‧‧Oxide film

12‧‧‧Sea of Epoxy Compounds

13‧‧‧ Island of acrylic resin

21‧‧‧ substrate

22‧‧‧Wiring pattern

23‧‧‧Lighting element

24‧‧‧n electrode

25‧‧‧p electrode

26‧‧‧Bump

30‧‧‧Anisotropic conductive film

50‧‧‧Anisotropic conductive adhesive

51‧‧‧Wiring board

52‧‧‧LED chip

53‧‧‧Hot press tool

54‧‧‧Tool

55‧‧‧Test roller

61‧‧‧Aluminum nitride particles

62‧‧‧Cu particles

63‧‧‧ Diamond particles

FIG. 1 is a cross-sectional view of a sea island model when an epoxy compound is used as a sea and an acrylic resin is used as an island.

2 is a cross-sectional view illustrating solder particles.

3 is a cross-sectional view showing an example of a light-emitting device.

Fig. 4 is a cross-sectional view showing the outline of a 90-degree peel strength test.

FIG. 5 is a diagram for explaining the manufacturing steps of the LED mounting samples.

6 is a cross-sectional view showing the outline of a wafer shear strength test.

FIG. 7 is a diagram illustrating the case of using diamond particles as a material for heat dissipation.

FIG. 8 is a diagram illustrating the case of using copper powder as a material for heat dissipation.

FIG. 9 is a diagram illustrating the case of using aluminum nitride powder as a material for heat dissipation.

FIG. 10 is a diagram illustrating the heat dissipation characteristics of the resin adhesive.

Fig. 11 is a diagram for explaining a bending test.

Fig. 12 is a diagram for explaining a bending test.

Hereinafter, referring to the drawings, an embodiment of the present invention (hereinafter, referred to as the present embodiment) will be described in detail in the following order. Furthermore, the present invention is not limited to the following embodiments, and of course various changes can be made without departing from the gist of the present invention. In addition, the figures are schematic, and the ratio of each size may differ from the actual one. Specific dimensions, etc. should be judged taking into account the following description. Also, of course, the schema also includes each other The size relationship or ratio is different.

1. Adhesive

2. Connection structure

3. Examples

<1. Adhesive>

Adhesives applicable to the present invention contain an alicyclic epoxy compound or a hydrogenated epoxy compound, a cationic catalyst, an acrylic resin with a weight average molecular weight of 50,000 to 900,000, and solder particles, and the acrylic resin contains 0.5 to 10% by weight of acrylic acid, And 0.5~10wt% of acrylate with hydroxyl group.

FIG. 1 is a cross-sectional view of a sea-island model when an epoxy compound is sea and an acrylic resin is an island at the interface between an adhesive and an oxide film. The sea island model is a model of a cured product in which the island 13 of acrylic resin dispersed in the sea 12 of epoxy compound is in contact with the oxide film 11a of the wiring 11.

In the cured product model, acrylic acid in the acrylic resin reacts with the epoxy compound to produce a connection between the island 13 of the acrylic resin and the sea 12 of the epoxy compound, and roughens the surface of the oxide film 11a to strengthen the epoxy compound The investment anchor effect of the sea of 12. In addition, the acrylate having a hydroxyl group in the acrylic resin obtains the electrostatic adhesion to the wiring 11 due to the polarity of the hydroxyl group. By bonding the oxide film 11 a with the cured product of the island 13 of acrylic resin and the sea 12 of epoxy compound as described above, excellent adhesion can be obtained.

Next, the solder particles will be described. Specifically, an example will be described in which an LED element is bonded to an aluminum wiring substrate having a wiring pattern whose surface is an oxide with an adhesive. Fig. 2 is a cross-sectional view illustrating the action of solder particles contained in an adhesive.

As shown in FIG. 2, the solder particles 1 are added to the resin binder 3 configured as described above together with the conductive particles 2 described below. The solder particles 1 are dispersed together with the conductive particles 2 between the electrode 10 of the LED element and the wiring 11 of the aluminum wiring board, and are melted in the pressure bonding step to become the molten solder 1a.

Here, the electrode 10 of the LED element is composed of Au or Au-Sn. The solder particles 1 melt when heated above the melting point, and solidify into a substantially columnar shape when cooled below the freezing point, and one end face 1b is metal-bonded to the electrode 10. On the other hand, the solder particles 1 cannot be combined with the metal of the wiring 11. The reason for this is that there is an oxide film 11a formed of aluminum oxide on the wiring 11, and in the normal pressure bonding step, the solder 1a is melted and the wiring 11 of the aluminum wiring board cannot be metal-bonded. Therefore, there is no case where the melting solder 1a contributes to electrical conduction between the electrode 10 and the wiring 11 of the LED element.

However, the melted solder 1a is metal-bonded to the electrode 10 at the end surface 1b, so the electrode 10 and the melted solder 1a form a structure. As a result, the adhesion between the LED element and the adhesive increases. Specifically, when there is no molten solder 1a, the electrode 10 of the LED element and the adhesive are only in two-dimensional contact, but the structure formed by the electrode 10 of the LED element and the molten solder 1a has a three-dimensional structure, so As a result, the area between the electrode 10 and the adhesive increases. That is, the molten solder 1a is bonded to a part of the metal of the electrode 10, thereby acting as a pile (anchor) to the adhesive, so that the bonding strength can be improved between the electrode 10 and the adhesive.

In addition, since the molten solder 1a is metal-bonded with the electrode 10, it is not a point contact like other particles used as a heat dissipation material, but a surface contact, and heat can be dissipated from the LED element side through the molten solder 1a, which can be leapfrogged. Improve heat dissipation characteristics. In addition, the contact surface with the wiring 11 also interposes the oxide film 11a, but it becomes a surface contact and becomes easy to transfer heat. In this respect, Can also improve the heat dissipation characteristics. In addition, the comparison with other materials for heat dissipation is explained in detail in the comparative example.

The solder particles 1 can be selected from, for example, Sn-Pb series, Pb-Sn-Sb series, Sn-Sb series, Sn-Pb-Bi series, Bi-Sn series specified in JIS Z 3282-1999 according to electrode materials or connection conditions, etc. , Sn-Cu system, Sn-Pb-Cu system, Sn-In system, Sn-Ag system, Sn-Pb-Ag system, Pb-Ag system, etc. are selected as appropriate. In addition, the shape of the solder particles 1 can be appropriately selected from granular, scaly, and the like.

Furthermore, the average particle diameter (D50) of the solder particles 1 is preferably 3 μm or more and less than 30 μm, and the added amount of the solder particles 1 is preferably 50 parts by mass or more and less than 150 parts by mass. The reason is that if the added amount is too small, the anchoring effect as described above cannot be expected, and if the added amount is excessively increased, the resin binder 3 relatively decreases, and the adhesion as an adhesive decreases.

In addition, the melting point of the solder particles 1 is preferably the one below the assembly temperature. If the solder particles 1 having such a melting point are used, the solder particles 1 can be melted by heating during assembly (crimping step), so there is no need to add a heating step for melting the solder particles 1 only. That is, the adhesive can be hardened and the solder particles 1 can be melted. The reason for this is that in order to form the molten solder 1a, excessive heating stress is not applied to the LED element or the substrate. For example, in a case where a resin substrate using aluminum wiring is attached to an LED element, the heat resistance of the resin substrate is taken into consideration at 180°C. Therefore, in this case, it is preferably 180°C or lower.

Next, as the alicyclic epoxy compound, those having two or more epoxy groups in the molecule can be preferably exemplified. The alicyclic epoxy compound may be liquid or solid. Specifically, 3,4-epoxycyclohexenylmethyl-3',4'-epoxycyclohexene carboxylate, epoxy Hexahydrobisphenol A, etc. Among these, 3,4-epoxycyclohexene can be preferably used in terms of ensuring light transmittance suitable for the construction of LED elements and the like and excellent rapid hardening property in the cured product. Methyl-3',4'-epoxycyclohexene carboxylate.

As the hydrogenated epoxy compound, a known hydrogenated epoxy compound such as the hydride of the alicyclic epoxy compound or the bisphenol A type and the bisphenol F type can be used.

The alicyclic epoxy compound or hydrogenated epoxy compound may be used alone or in combination of two or more. In addition to these epoxy compounds, other epoxy compounds may be used in combination as long as the effects of the present invention are not impaired. For example, bisphenol A, bisphenol F, bisphenol S, tetramethyl bisphenol A, diaryl bisphenol A, hydroquinone, catechol, resorcinol, cresol, Tetrabromobisphenol A, trihydroxybiphenyl, benzophenone, bisresorcinol, bisphenol hexafluoroacetone, tetramethylbisphenol A, tetramethylbisphenol F, tris(hydroxyphenyl)methane, Glycidyl ether obtained from the reaction of polyphenols such as bixylenol, phenolic novolak, cresol novolak and epichlorohydrin; glycerin, neopentyl glycol, ethylene glycol, propylene glycol, butylene glycol, hexanediol Polypropylene oxide obtained by reacting aliphatic polyhydric alcohols such as alcohol, polyethylene glycol, polypropylene glycol and epichlorohydrin; reacting hydroxycarboxylic acids such as p-hydroxybenzoic acid and β-hydroxynaphthoic acid with epichlorohydrin The obtained glycidyl ether ester; free phthalic acid, methyl phthalic acid, isophthalic acid, terephthalic acid, tetrahydrophthalic acid, internal methylenetetrahydrophthalic acid, internal Polypropylene oxide obtained from methylenehexahydrophthalic acid, 1,2,4-benzenetricarboxylic acid, polycarboxylic acid polymerized fatty acid; epoxypropylamine obtained from aminophenol, aminoalkanol Glycidyl ether; glycidyl propyl glycidyl ester obtained from aminobenzoic acid; from aniline, toluidine, tribromoaniline, xylylenediamine, diaminocyclohexane, diaminomethyl Cyclopropylamine, 4,4'-diaminodiphenylmethane, 4,4'-diaminodiphenylbenzene, etc., obtained from glycidylamine; epoxidized polyolefin and other known epoxy resins.

Examples of the cationic catalyst include aluminum chelate latent hardeners and Latent cationic hardeners such as azole-based latent hardeners and lacquer-based latent hardeners. Among these, an aluminum chelate latent hardener excellent in rapid hardening property can be preferably used.

The content of the cationic catalyst has a tendency that if it is too little, the reactivity disappears, and if it is too much, the product life of the adhesive decreases, so it is preferably 0.1 to 30 parts by mass, more preferably 0.5 with respect to 100 parts by weight of the epoxy compound. ~20 parts by mass.

The weight average molecular weight of the acrylic resin system is 50,000 to 900,000. In the cured product model shown in FIG. 1, the weight average molecular weight of acrylic resin has an influence on the size of the island 13 of acrylic resin. By the weight average molecular weight of acrylic resin being 50000~900000, the acrylic resin of appropriate size can be made The island 13 is in contact with the oxide film 11a. When the weight average molecular weight of the acrylic resin is less than 50000, the area of contact between the island 13 of the acrylic resin and the oxide film 11a becomes small, and the effect of improving the adhesion cannot be obtained. In addition, when the weight average molecular weight of the acrylic resin exceeds 900,000, the island 13 of the acrylic resin becomes larger, and it cannot be said that the oxide film 11a is bonded to the entire cured product of the island 13 of the acrylic resin and the sea of the epoxy compound 12 , And then the force drops.

In addition, the acrylic resin contains 0.5 to 10 wt% of acrylic acid, more preferably 1 to 5 wt%. By including acrylic resin in an amount of 0.5 to 10 wt% in acrylic resin, the connection between the island of acrylic resin 13 and the sea of epoxy compound 12 is generated due to the reaction with the epoxy compound, and the surface of the oxide film 11a becomes roughened to strengthen and ring The anchor effect of the sea of oxygen compounds 12.

In addition, the acrylic resin contains 0.5 to 10 wt% of acrylate having a hydroxyl group, more preferably 1 to 5 wt%. By including 0.5 to 10 wt% of acrylate in the acrylic resin, the electrostatic adhesion to the wiring 11 is obtained due to the polarity of the hydroxyl group.

Examples of the acrylate having a hydroxyl group include 2-hydroxyethyl methacrylate Ester, 2-hydroxypropyl methacrylate, 2-hydroxyethyl acrylate, 2-hydroxypropyl acrylate, etc. Among these, 2-hydroxyethyl methacrylate excellent in adhesion to an oxide film can be preferably used.

In addition, acrylic resins include acrylates that do not have hydroxyl groups in addition to acrylic acid and acrylates that have hydroxyl groups. Examples of acrylates having no hydroxyl group include butyl acrylate, ethyl acrylate, and acrylic nitrile.

In addition, the content of the acrylic resin is preferably 1 to 10 parts by mass, and more preferably 1 to 5 parts by mass relative to 100 parts by mass of the epoxy compound. When the content of the acrylic resin is 1 to 10 parts by mass relative to 100 parts by mass of the epoxy compound, a cured product in which the island 13 of the acrylic resin is dispersed in the sea 12 of the epoxy resin with good density can be obtained.

In addition, the adhesive used in the present invention is intended to improve the adhesion with the inorganic material at the interface. As other components, it may further contain a silane coupling agent. Examples of the silane coupling agent include epoxy-based, methacryloxy-based, amine-based, ethylene-based, mercapto-sulfur-based, and ureide-based. These can be used alone or two or more types can be mixed. And use. Among these, the epoxy-based silane coupling agent can be preferably used in this embodiment.

In addition, in order to control the fluidity and improve the particle capture rate, the adhesive may contain an inorganic filler. The inorganic filler is not particularly limited, and silicon oxide, talc, titanium oxide, calcium carbonate, magnesium oxide, and the like can be used. Such an inorganic filler can be suitably used for the purpose of easing the stress of the connection structure connected by an adhesive. In addition, softeners such as thermoplastic resins and rubber components can also be blended.

According to such an adhesive, it is possible to obtain a high adhesion to difficult metals such as aluminum.

In addition, the adhesive may be an anisotropic conductive adhesive containing conductive particles. Make As the conductive particles, known conductive particles can be used. For example, particles of various metals or metal alloys such as nickel, iron, copper, aluminum, tin, lead, chromium, cobalt, silver, gold, etc., metal oxides, particles of carbon, graphite, glass, ceramics, plastics, etc. Those coated with metal on the surface, those coated with an insulating film on the surface of these particles, etc. In the case of a metal coated on the surface of the resin particles, as the resin particles, for example, epoxy resin, phenol resin, acrylic resin, acrylonitrile-styrene (AS) resin, benzoguanamine resin, dimethacrylate Particles of vinylbenzene resin, styrene resin, etc.

The average particle diameter of the conductive particles is usually 1 to 10 μm, more preferably 2 to 6 μm. Further, the average particle density is then based agent component of the conductive particles on the connection reliability and insulation reliability of the viewpoint, preferably 1,000 to 100,000 / mm ^2, more preferably 30,000 to 80,000 / mm ^2. Here, the content of the conductive particles is preferably 1 to 20 parts by mass.

According to such anisotropic conductive adhesive, excellent connection reliability can be obtained for aluminum wiring or ITO wiring having an oxide film.

<2. Connection structure>

Next, the connection structure to which the present invention is applied will be described. 3 is a cross-sectional view of an LED element as an electronic component that generates heat as an example of a connection structure. The connection structure includes a substrate 21 having a wiring pattern 22, an anisotropic conductive film 30 formed on an electrode of the wiring pattern 22, and a light-emitting element 23 structured on the anisotropic conductive film 30, and the anisotropic conductive film 30 is composed of the cured product of the anisotropic conductive adhesive. The light-emitting device is obtained by applying the above-mentioned anisotropic conductive adhesive between the wiring pattern 22 on the substrate 21 and the connection bumps 26, and flip-chip mounting the substrate 21 and the light-emitting element 23; The connection bumps 26 are formed on the n electrode 24 and the p electrode 25 of the LED element as the light emitting element 23, respectively.

In addition, the bump 26 described here is formed by plating Au or Au-Sn alloy. Therefore, the bump 26 corresponds to the electrode 10 described in FIG. 2, and the solder particles 1 are metal-bonded to the bump 26.

In this embodiment, the anisotropic conductive adhesive described above is used, whereby a substrate having a wiring pattern composed of aluminum can be preferably used. With this, the cost of LED products can be reduced.

In addition, a transparent substrate having a wiring pattern composed of a transparent conductive film such as ITO can be preferably used. In this way, for example, an LED element can be mounted on a transparent resin substrate with ITO (Indium Tin Oxide) wiring formed on a PET (Polyethylene terephthalate) substrate.

Furthermore, if necessary, a transparent mold resin with good heat dissipation characteristics may be used to seal the entire light-emitting element 23. In addition, a light reflection layer may be provided on the light emitting element 23. In addition, as the light-emitting element, in addition to the LED element, a known heat-generating electronic component can be used within a range that does not impair the effects of the present invention.

<3. Example> [Example] [First embodiment]

Hereinafter, the first embodiment of the present invention will be described. In this embodiment, various anisotropic conductive adhesives are produced, and LED components are fabricated by using these anisotropic conductive adhesives to mount LED elements on a substrate, and alloys of terminal portions with and without LED elements and solder particles are formed , Thermal resistance, and evaluation of aluminum adhesion. Furthermore, the invention is not limited to these embodiments.

[Measurement of Peel Strength]

Apply anisotropy to a white plate made of ceramics with a thickness of 100 μm The conductive adhesive was thermocompression-bonded to an aluminum sheet of 1.5 mm×10 mm under the conditions of 180° C.-1.5 N-30 sec, thereby producing a bonded body.

As shown in FIG. 4, using a tensile tester, the aluminum sheet of the bonded body was peeled in the Y-axis direction of 90° at a tensile speed of 50 mm/sec, and the maximum value of peel strength required for the peeling was measured.

[Production of LED construction samples]

As shown in Fig. 5, a sample of the LED structure was fabricated. A plurality of wiring substrates with a pitch of 50 μm (Al wiring of 50 μm—PI (polyimide) layer of 50 μm—Al base of 50 μm) are arranged on the platform, and an anisotropy of about 10 μg is coated on each wiring substrate 51 Conductive adhesive 50. On the anisotropic conductive adhesive 50, an LED chip (trade name: DA3547, maximum rating: 150mA, size: 0.35mm×0.46mm) 52 manufactured by Cree is mounted, and a flip chip is mounted using a hot press tool 53 , So as to obtain LED construction samples.

[Measurement of wafer shear strength]

As shown in FIG. 6, using a wafer shear tester, the bonding strength of each LED mounting sample was measured under the conditions of the shear speed of the tool 54 at 20 μm/sec and 25°C.

[Evaluation of alloy formation]

Using a microscope (SEM: Scanning Electron Microscope), etc., confirm whether the appearance of each LED mounting sample has been alloyed between the electrode part of the LED element and the solder particles. Specifically, if alloy formation is achieved, the electrode portion and the solder particles are in surface contact by melting the solder. Therefore, whether the alloy is formed can be judged by observing the expanded area of the molten solder, that is, whether the metal bonding has been achieved can be judged.

[Evaluation of Thermal Resistance]

Using a transient thermal resistance measuring device (manufactured by CATS Electronic Design) to measure the LED structure The thermal resistance value (℃/W). The measurement conditions were performed with If=200mA (constant current control).

[Overview]

If the terminal part of the LED element is formed with an alloy of solder particles, the thermal resistance value is "○", the peel strength is 2.0N or more, and the wafer shear strength is 5.0N or more, it is evaluated as "OK", except for this Outsiders are evaluated as "NG".

[Example 1]

30 parts by mass of solder particles with a melting point of 150°C and 10 parts by mass of conductive particles (product name: AUL704, manufactured by Sekisui Chemical Industry Co., Ltd.) are dispersed in 100 parts of an alicyclic epoxy compound (product name: Celloxide2021P, manufactured by Daicel Chemical Company) 100 Mass parts, latent cationic hardener (aluminum chelate-based latent hardener) 5 parts by mass, acrylic resin (butyl acrylate (BA): 15%, ethyl acrylate (EA): 63%, acrylonitrile (AN ): 20%, acrylic acid (AA): 1% by weight, 2-hydroxyethyl methacrylate (HEMA): 1% by weight, weight average molecular weight Mw: 700,000) 3 parts by mass of an adhesive composed of anisotropic conductive Then agent. In addition, the curing conditions at the time of manufacturing the LED structure sample were set to 180°C-1.5N-30sec.

In addition, in each Example, the average particle diameter of the solder particle was 5 micrometers, 7 micrometers, 10 micrometers, 12 micrometers, and 25 micrometers. No significant difference was found in the particle size within the above range, so the test results for each particle size were omitted, and the results of the examples of the present application can be obtained by using a particle size of at least the above range. The same applies to the following examples and comparative examples prepared with solder particles.

Table 1 shows the evaluation results of Example 1. It was confirmed that the alloy was formed, the thermal resistance value was 17 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Example 2]

An anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 150°C and the formulation was 40 parts by mass.

Table 1 shows the evaluation results of Example 2. It was confirmed that the alloy was formed, the thermal resistance value was 16 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Example 3]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 150°C and the formulation was 60 parts by mass.

Table 1 shows the evaluation results of Example 3. It was confirmed that the alloy was formed, the thermal resistance value was 16 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Example 4]

An anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 150°C and the formulation was 80 parts by mass.

Table 1 shows the evaluation results of Example 4. It was confirmed that the alloy was formed, the thermal resistance value was 15 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Example 5]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 170°C and the formulation was 30 parts by mass.

Table 1 shows the evaluation results of Example 5. Confirmation of alloy formation, The thermal resistance value was 16 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Example 6]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 170°C and the formulation was 80 parts by mass.

Table 1 shows the evaluation results of Example 6. It was confirmed that the alloy was formed, the thermal resistance value was 16 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is OK.

[Comparative Example 1]

An anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the solder particles were not prepared.

Table 1 shows the evaluation results of Comparative Example 1. No alloy formation was confirmed, the thermal resistance value was 29 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 2]

An anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 150°C and the formulation was 160 parts by mass.

Table 1 shows the evaluation results of Comparative Example 2. It was confirmed that the alloy was formed, the thermal resistance value was 16 (K/W), and the peel strength was 1.2N. In addition, the wafer shear strength of the LED structure sample was 2.0N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 3]

The melting point of the solder particles is set to 170°C, and the formulation is set to 160 parts by mass. In Example 1, an anisotropic conductive adhesive was produced in the same manner.

Table 1 shows the evaluation results of Comparative Example 3. It was confirmed that the alloy was formed, the thermal resistance value was 16 (K/W), and the peel strength was 1.2N. In addition, the wafer shear strength of the LED structure sample was 2.0N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 4]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 200° C. and the formulation was 30 parts by mass.

Table 1 shows the evaluation results of Comparative Example 4. No alloy formation was confirmed, the thermal resistance value was 26 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 5]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1, except that the melting point of the solder particles was 200°C and the formulation was 80 parts by mass.

Table 1 shows the evaluation results of Comparative Example 5. No alloy formation was confirmed, the thermal resistance value was 23 (K/W), and the peel strength was 4.0N. In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 6]

The anisotropic conductive adhesive was prepared in the same manner as in Example 1 except that the melting point of the solder particles was 200° C. and the formulation was 160 parts by mass.

Table 1 shows the evaluation results of Comparative Example 6. No alloy formation was confirmed, the thermal resistance value was 23 (K/W), and the peel strength was 1.2N. In addition, the wafer shear strength of the LED structure sample was 2.0N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 7]

Instead of solder particles, 60 parts by mass of alumina powder having an average particle diameter of 0.4 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 7. No alloy formation was confirmed, and the thermal resistance value was 25 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.5N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 8]

Instead of solder particles, 150 parts by mass of alumina powder having an average particle diameter of 0.4 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 8. No alloy formation was confirmed, and the thermal resistance value was 23 (K/W). In addition, the wafer shear strength of the LED structure sample was 5.3N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 9]

Instead of solder particles, 60 parts by mass of alumina powder having an average particle diameter of 3 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 9. No alloy formation was confirmed, and the thermal resistance value was 29 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.8N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 10]

Instead of solder particles, 150 parts by mass of alumina powder having an average particle diameter of 3 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 10. No alloy formation was confirmed, and the thermal resistance value was 28 (K/W). In addition, the wafer shear strength of the LED structure sample was 6.2N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 11]

Instead of solder particles, 60 parts by mass of alumina powder having an average particle diameter of 10 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 11. No alloy formation was confirmed, and the thermal resistance value was 35 (K/W). In addition, the wafer shear strength of the LED structure sample was 6.1N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 12]

Instead of solder particles, 150 parts by mass of alumina powder having an average particle diameter of 10 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 2 shows the evaluation results of Comparative Example 12. No alloy formation was confirmed, and the thermal resistance value was 33 (K/W). In addition, the wafer shear strength of the LED structure sample was 5.5N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 13]

Instead of solder particles, aluminum nitride powder with an average particle size of 1.5 μm as a heat dissipation material 60 parts by mass was blended in the resin binder, except that the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 13. No alloy formation was confirmed, and the thermal resistance value was 22 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.1N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 14]

Instead of solder particles, 150 parts by mass of aluminum nitride powder having an average particle diameter of 1.5 μm as a heat dissipating material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 14. No alloy formation was confirmed, and the thermal resistance value was 19 (K/W). In addition, the wafer shear strength of the LED structure sample was 5.9N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 15]

Instead of solder particles, 60 parts by mass of Ni powder having an average particle diameter of 3 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 13. No alloy formation was confirmed, and the thermal resistance value was 28 (K/W). In addition, the wafer shear strength of the LED structure sample was 7.9N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 16]

Instead of solder particles, 150 parts by mass of Ni powder having an average particle diameter of 3 μm as a heat-dissipating material was blended into the resin binder, and the anisotropic conductive material was produced in the same manner as in Example 1. Then agent.

Table 3 shows the evaluation results of Comparative Example 16. No alloy formation was confirmed, and the thermal resistance value was 27 (K/W). In addition, the wafer shear strength of the LED structure sample was 6.0N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 17]

Instead of solder particles, 60 parts by mass of Cu powder having an average particle diameter of 10 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 17. No alloy formation was confirmed, and the thermal resistance value was 41 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.12N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 18]

Instead of solder particles, 150 parts by mass of Cu powder having an average particle diameter of 10 μm as a heat dissipation material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 18. No alloy formation was confirmed, and the thermal resistance value was 38 (K/W). In addition, the wafer shear strength of the LED structure sample was 6.2N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 19]

Instead of solder particles, 60 parts by mass of diamond powder having an average particle diameter of 0.3 μm as a heat dissipating material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 19. No alloy formation was confirmed, and the thermal resistance value was 21 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.3N. Therefore, the comprehensive evaluation is NG.

[Comparative Example 20]

Instead of solder particles, 150 parts by mass of diamond powder having an average particle diameter of 0.3 μm as a heat dissipating material was blended into the resin binder, and the anisotropic conductive adhesive was prepared in the same manner as in Example 1.

Table 3 shows the evaluation results of Comparative Example 20. No alloy formation was confirmed, and the thermal resistance value was 22 (K/W). In addition, the wafer shear strength of the LED structure sample was 8.1N. Therefore, the comprehensive evaluation is NG.

In Comparative Example 1, solder particles are not blended, so metal bonding by melting solder does not occur, and the thermal resistance value becomes too large, thereby deteriorating heat dissipation characteristics.

In addition, in Comparative Examples 2 and 3, there are many solder particles, so although the molten solder is formed, the adhesion between the aluminum wiring substrate and the anisotropic conductive adhesive is reduced, and the anisotropic conductive adhesive and the LED element Then the force drops.

Also, in Comparative Examples 4, 5, and 6, the melting point of the solder particles is set at 200°C, so the solder will not be sufficiently melted in the crimping step, and the metal bonding by the melted solder will not occur, resulting in anisotropy The adhesive force between the conductive adhesive and the LED element decreases, and the thermal resistance value becomes too large, so that the heat dissipation characteristics deteriorate.

In addition, Comparative Examples 7, 8, 9, 10, 11, and 12 use alumina powder as a heat dissipation material, so there is no metal bonding by melting solder, but the adhesion between the anisotropic conductive adhesive and the LED element When it drops, the thermal resistance value becomes too large, and the heat dissipation characteristics become poor. The thermal conductivity of the alumina powder is 40 W/mK, but according to comparison with the examples of the present application, even if it is contained in the adhesive instead of the solder particles, the desired characteristics cannot be obtained.

In addition, Comparative Examples 13 and 14 use aluminum nitride powder as a heat dissipation material, so there is no metal bonding by melting solder, and the adhesion between the anisotropic conductive adhesive and the LED element is reduced, and the thermal resistance value It becomes too large, so that the heat dissipation characteristics deteriorate. The thermal conductivity of aluminum nitride powder is 180 W/mK, but according to comparison with the examples of the present application, even if it is contained in the adhesive instead of the solder particles, the desired characteristics cannot be obtained.

Here, the case of adding aluminum nitride as a heat dissipation material will be discussed. As shown in FIG. 7, when the aluminum nitride particles 61 are added to the resin binder 3, there is no case of melting as easily as solder particles, so the particle shape is maintained, and the electrode 10 and the aluminum nitride particles 61 become Point of contact. Therefore, the area for transferring heat from the LED element becomes very small, and the heat dissipation characteristics become worse compared to the case where solder particles are used. In addition, the contact between the aluminum nitride particles 61 and the wiring 11 also becomes a point contact. Therefore, the heat dissipation characteristics from the aluminum nitride particles 61 to the wiring substrate side also deteriorate.

In addition, Comparative Examples 15 and 16 use Ni powder as a heat dissipation material, so there is no metal bonding by melting solder, and the adhesion between the anisotropic conductive adhesive and the LED element decreases, and the thermal resistance value becomes If it is too large, the heat dissipation characteristics will deteriorate. The thermal conductivity of Ni powder is 95 W/mK, but according to comparison with the examples of the present application, even if it is contained in the adhesive instead of the solder particles, the desired characteristics cannot be obtained.

In addition, Comparative Examples 17 and 18 use Cu powder as a heat dissipation material, so metal bonding by melting solder does not occur, and the adhesion between the anisotropic conductive adhesive and the LED element decreases, and the thermal resistance value becomes If it is too large, the heat dissipation characteristics will deteriorate. The thermal conductivity of Cu powder is 400 W/mK, but according to comparison with the examples of the present application, even if it is contained in the adhesive instead of the solder particles, the desired characteristics cannot be obtained.

Here, the case of adding Cu particles as a heat dissipation material will be discussed. As shown in FIG. 8, when the Cu particles 62 are added to the resin binder 3, it is the same as the aluminum nitride particles 61 in the following respects: there is no case of melting as easily as solder particles, so the particle shape is maintained, The electrode 10 comes into point contact with the Cu particles 62. In addition, the particle diameter of the Cu particles 62 is very large, so the thickness of the adhesive becomes thicker. Even if Cu particles with high thermal conductivity are used, the thickness of the adhesive layer as a whole hinders the heat dissipation characteristics, so that the desired heat dissipation characteristics cannot be obtained.

In addition, Comparative Examples 19 and 20 use diamond powder as a heat dissipation material, so metal bonding by melting solder does not occur, and the adhesion between the anisotropic conductive adhesive and the LED element decreases, and the thermal resistance value becomes If it is too large, the heat dissipation characteristics will deteriorate. Diamond powder The thermal conductivity is 1500W/mK, but according to comparison with the examples of the present application, even if it is contained in the adhesive instead of the solder particles, the desired characteristics cannot be obtained.

Here, the case of adding diamond particles as a heat dissipation material will be discussed. As shown in FIG. 9, when the diamond particles 63 are added to the resin binder 3, the diamond particles 63 are smaller than the thickness of the adhesive layer, so they cannot contact the electrode portion of the LED element or the wiring on the substrate side. That is, no heat transfer path is formed from the LED element to the wiring substrate side, so even if diamond particles with high thermal conductivity are used, the desired heat dissipation characteristics cannot be obtained.

On the other hand, Examples 1 to 6 are formulated with an alicyclic epoxy compound, a latent cationic hardener, and an acrylic resin having acrylic acid (AA) and 2-hydroxyethyl methacrylate (HEMA), so it has optical The characteristics of the application, in addition, can obtain high adhesion and excellent conduction reliability for aluminum wiring with an oxide film, and the melting point of the solder particles is set below the assembly temperature, so the solder particles are melted in the pressure bonding step The melting solder is combined with the electrode metal of the LED element to obtain higher adhesion and excellent heat dissipation characteristics.

In addition, in FIG. 10, the heat dissipation characteristic of the resin adhesive is shown as a reference. Resin A is an example where the thermal conductivity is adjusted to 10W/mK, Resin B is an example where the thermal conductivity is adjusted to 30W/mK, Resin C is an example where the thermal conductivity is adjusted to 50W/mK, and Resin D is a thermal example An example of conductivity adjustment to 70W/mK. Generally, it is known that if the volume ratio (vol%) of the heat-dissipating resin in the adhesive layer is not high, the heat dissipation characteristics cannot be obtained. The thermal resistance is defined by the layer thickness/(adjacent area×thermal conductivity). Therefore, if the layer thickness is too large, the thermal resistance becomes high. Therefore, it can be seen that the heat dissipation material with a larger particle size increases the layer thickness, which is not good.

[Second Embodiment]

Hereinafter, the second embodiment of the present invention will be described. In this embodiment, various Anisotropic conductive adhesive, using these anisotropic conductive adhesives to mount an LED element on a substrate to prepare an LED structure sample, and evaluated the adhesive strength and on-resistance. Furthermore, the invention is not limited to these embodiments.

[Measurement of Peel Strength]

On a white plate made of ceramics, anisotropic conductive adhesive is applied with a thickness of 100 μm, and an aluminum sheet of 1.51 nm×10 mm is thermocompression-bonded under the conditions of 180° C.-1.5 N-30 sec to produce a bonded body .

As shown in FIG. 4, using a tensile tester, the aluminum sheet of the bonded body was peeled in the Y-axis direction of 90° at a tensile speed of 50 mm/sec, and the maximum value of peel strength required for the peeling was measured.

[Production of LED construction samples]

As shown in Fig. 5, a sample of the LED structure was fabricated. A plurality of wiring substrates with a pitch of 50 μm (Al wiring of 50 μm—PI (polyimide) layer of 50 μm—Al base of 50 μm) are arranged on the platform, and an anisotropy of about 10 μg is coated on each wiring substrate 51 Conductive adhesive 50. On the anisotropic conductive adhesive 50, an LED chip (trade name: DA3547, maximum rating: 150mA, size: 0.35mm×0.46mm) 52 manufactured by Cree is mounted, and a flip chip is mounted using a hot press tool 53 , So as to obtain LED construction samples.

[Measurement of wafer shear strength]

As shown in FIG. 6, using a wafer shear tester, the bonding strength of each LED mounting sample was measured under the conditions of the shear speed of the tool 54 at 20 μm/sec and 25°C.

[Evaluation of on-resistance before bending test]

The on-resistance of each LED configuration sample was measured at the initial stage and after the thermal cycle test (TCT). The hot and cold cycle test is to expose the LED configuration sample to -40℃ and 100℃ for 30 minutes each Then, 1000 cycles were performed as a cycle of cooling and heating, and the on-resistance was measured. The evaluation of the on-resistance is to measure the Vf value at If=50mA, and the degree of increase of the Vf value from the Vf value of the test score table is less than 5% is set to "○", and the case of more than 5% is set to "× ".

[Evaluation of on-resistance value after bending test]

As shown in FIGS. 11 and 12, after conducting a test in which each LED mounting sample was pressed against the side of the cylindrical test roller 55 to bend it, the on-resistance was measured. Specifically, the LED wafer 52 is substantially rectangular, so each of the following tests is performed: the test of winding the wiring board 51 in the longitudinal direction (X-axis direction) of the LED wafer 52 as shown in FIG. 11; And as shown in FIG. 12, a test of winding the wiring board 51 in the short side direction (Y-axis direction) of the LED wafer 52.

In addition, the smaller the radius (R) of the test roller 55, the stronger the bending stress exerted on each LED mounting sample. Therefore, a plurality of diameters are used for the test. Specifically, the test roller 55 was subjected to a bending test using a diameter of 20 mm (R=10 mm), a diameter of 10 mm (R=5 mm), and a diameter of 6 mm (R=3 mm), and the on-resistance was measured for each. The evaluation of the on-resistance is to measure the Vf value at If=50mA, and the degree of increase of the Vf value from the Vf value of the test score table is less than 5% is set to "○", and the case of more than 5% is set to "× ".

Here, in the case where the adhesion of the anisotropic conductive adhesive to the wiring substrate is weak, the conduction characteristics disappear after the bending test, that is, the on-resistance value increases. The reason for this is that, in the case where the adhesion force is weak, there is a case where a distance is generated between the member between the electrode and the wiring due to the bending test, and the contact of the conductive particles disappears.

[Overview]

The on-resistance values after the initial and after the thermal cycle test are both "○", the peel strength is 2.0 N or more, the wafer shear strength is 5.0 N or more, and the continuity evaluation after the bending test is "○" The person evaluated as "OK", and the others were evaluated as "NG".

[Example 7]

30 parts by mass of solder particles with a melting point of 150°C and 10 parts by mass of conductive particles (product name: AUL704, manufactured by Sekisui Chemical Industry Co., Ltd.) are dispersed in an alicyclic epoxy compound (product name: Celloxide2021P, Daicel) 100 parts by mass of chemical company, 5 parts by mass of latent cationic hardener (aluminum chelate-based latent hardener), acrylic resin (butyl acrylate (BA): 15%, ethyl acrylate (EA): 63% , Acrylonitrile (AN): 20%, Acrylic acid (AA): 1wt%, 2-hydroxyethyl methacrylate (HEMA): 1wt%, weight average molecular weight Mw: 700,000) 3 parts by mass of the adhesive To produce anisotropic conductive adhesive. In addition, the curing conditions at the time of manufacturing the LED structure sample were set to 180°C-1.5N-30sec.

In addition, in each Example, the average particle diameter of the solder particle was 5 micrometers, 7 micrometers, 10 micrometers, 12 micrometers, and 25 micrometers. No significant difference was found in the particle size within the above range, so the test results for each particle size were omitted, and the results of the examples of the present application can be obtained by using a particle size of at least the above range. The same applies to the following examples and comparative examples prepared with solder particles.

Table 4 shows the evaluation results of Example 7. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is ○ when it is 10 mm and ○ when it is 6 mm in diameter. Therefore, the comprehensive evaluation is OK.

[Example 8]

The melting point of the solder particles is set to 150°C, and the formulation is set to 80 parts by mass. In Example 7, an anisotropic conductive adhesive was produced in the same manner.

Table 4 shows the evaluation results of Example 8. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is ○ when it is 10 mm and ○ when it is 6 mm in diameter. Therefore, the comprehensive evaluation is OK.

[Example 9]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 170°C and the formulation was 30 parts by mass.

Table 4 shows the evaluation results of Example 9. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is ○ when it is 10 mm and ○ when it is 6 mm in diameter. Therefore, the comprehensive evaluation is OK.

[Example 10]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 170°C and the formulation was 80 parts by mass.

Table 4 shows the evaluation results of Example 10. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is ○ when it is 10 mm and ○ when it is 6 mm in diameter. Therefore, the comprehensive evaluation is OK.

[Comparative Example 21]

An anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the solder particles were not prepared.

Table 4 shows the evaluation results of Comparative Example 21. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 22]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 150°C and the formulation was 160 parts by mass.

Table 4 shows the evaluation results of Comparative Example 22. The initial peel strength was 1.2N. In addition, the initial wafer shear strength of the LED assembly sample was 2.0N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 23]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 170°C and the formulation was 160 parts by mass.

Table 4 shows the evaluation results of Comparative Example 23. The initial peel strength was 1.2N. In addition, the initial wafer shear strength of the LED assembly sample was 2.0N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, and after 1000 cycles of the cold and hot cycle test The conductivity evaluation was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm, × when the diameter was 10 mm, and × when the diameter was 6 mm. Therefore, the comprehensive evaluation is NG.

[Comparative Example 24]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 200° C. and the formulation was 30 parts by mass.

Table 4 shows the evaluation results of Comparative Example 24. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 25]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 200° C. and the formulation was 80 parts by mass.

Table 4 shows the evaluation results of Comparative Example 25. The initial peel strength was 4.0N. In addition, the initial wafer shear strength of the LED assembly sample was 8.5N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 26]

The anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that the melting point of the solder particles was 200°C and the formulation was 160 parts by mass.

Table 4 shows the evaluation results of Comparative Example 26. Initial peel strength It is 1.2N. In addition, the initial wafer shear strength of the LED assembly sample was 2.0N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ○, and the conduction evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 27]

Disperse 10 parts by mass of conductive particles (product name: AUL704, manufactured by Sekisui Chemical Co., Ltd.) in 50 parts by mass of an alicyclic epoxy compound (product name: Celloxide 2021P, manufactured by Daicel Chemical Company) as an adhesive B, and an anhydride hardener ( 40 parts by mass of methylhexahydrophthalic anhydride, 3 parts by mass of acrylic resin (BA: 15%, EA: 63%, AN: 20%, AA: 1wt%, HEMA: 1wt%, Mw: 200,000) Among the constructed adhesives, anisotropic conductive adhesives are produced. No solder particles are added. In addition, the curing conditions at the time of manufacturing the LED structure sample were 230°C-1.5N-30sec.

Table 5 shows the evaluation results of Comparative Example 27. The initial peel strength did not reach 0.5N. In addition, the initial wafer shear strength of the LED assembly sample was 3.8N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 28]

The anisotropic conductive adhesive was prepared in the same manner as in Comparative Example 27 except that the melting point of the solder particles was 170° C. and the formulation was added to the adhesive by 80 parts by mass.

Table 5 shows the evaluation results of Comparative Example 28. The initial peel strength did not reach 0.5N. In addition, the initial wafer shear strength of the LED assembly sample was 3.8N. Again, bending test The initial conductivity evaluation of the LED configuration sample before the inspection is ○, the conductivity evaluation after the 1000 cycle of the heat and cold cycle test is ×, and the conductivity evaluation after the bending test is × when the diameter of the test roller is 20 mm, and when the diameter is 10 mm It is × when the diameter is 6 mm. Therefore, the comprehensive evaluation is NG.

[Comparative Example 29]

An anisotropic conductive adhesive was prepared in the same manner as in Example 7 except that 100 parts by mass of cycloolefin was used as the binder C instead of the alicyclic epoxy compound. No solder particles are added. In addition, the curing conditions at the time of manufacturing the LED structure sample were set to 180°C-1.5N-240sec.

Table 5 shows the evaluation results of Comparative Example 29. The initial peel strength was 1.4N. In addition, the initial wafer shear strength of the LED assembly sample was 7.2N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 30]

The anisotropic conductive adhesive was prepared in the same manner as in Comparative Example 29 except that the melting point of the solder particles was 170° C. and the formulation was added to the adhesive by 80 parts by mass.

Table 5 shows the evaluation results of Comparative Example 30. The initial peel strength was 1.4N. In addition, the initial wafer shear strength of the LED assembly sample was 7.2N. In addition, the initial conductivity evaluation of the LED mounting sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was × when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 31]

Instead of an alicyclic epoxy compound, a bisphenol F-type epoxy compound is used as the binder D, so that An anisotropic conductive adhesive was produced in the same manner as in Example 7 except that the latent cationic hardener was replaced with an anionic hardener (amine-based hardener), and the acrylic resin was not formulated. No solder particles are added. In addition, the curing conditions at the time of manufacturing the LED mounting sample were set to 150°C-1.5N-30sec.

Table 5 shows the evaluation results of Comparative Example 31. The initial peel strength was 2.5N. In addition, the wafer shear strength at the initial stage of the LED assembly sample was 7.1N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

[Comparative Example 32]

The anisotropic conductive adhesive was prepared in the same manner as in Comparative Example 31 except that the melting point of the solder particles was 170° C. and the formulation was added to the adhesive by 80 parts by mass.

Table 5 shows the evaluation results of Comparative Example 32. The initial peel strength was 2.5N. In addition, the wafer shear strength at the initial stage of the LED assembly sample was 7.1N. In addition, the initial conductivity evaluation of the LED structure sample before the bending test was ○, the conductivity evaluation after the 1000-cycle cooling and heating test was ×, and the conductivity evaluation after the bending test was ○ when the diameter of the test roller was 20 mm. It is × when it is 10 mm, and it is × when it is 6 mm in diameter. Therefore, the comprehensive evaluation is NG.

In Comparative Example 21, the solder particles were not formulated, so the molten solder was not formed and the anchoring effect could not be obtained. Therefore, the adhesion between the anisotropic conductive adhesive and the LED element was reduced. After the bending test using a test roller with a diameter of 10 mm or less, The conduction reliability becomes low.

In addition, Comparative Examples 22 and 23 have a large amount of solder particles. Therefore, although molten solder is formed, the adhesion between the aluminum wiring substrate and the anisotropic conductive adhesive decreases, which results in the anisotropic conductive adhesive and the LED device. Then the bonding force decreases, and after the bending test, the conduction reliability becomes low.

In addition, the melting point of the solder particles of the comparative examples 24, 25, and 26 is set to 200°C, so the solder will not be sufficiently melted in the crimping step, and the metal bonding by the melted solder will not occur, thereby causing anisotropic conduction The adhesive force between the adhesive and the LED element decreased, and the conduction reliability became lower after the TCT test and after the bending test.

In Comparative Examples 27 and 28, acid anhydride was used as the hardener as the binder B. Therefore, regardless of the presence or absence of solder particles, the conduction reliability became lower after the TCT test and after the bending test. From this, it can be seen that there is an effect due to the combination of the adhesive A and the solder particles.

In Comparative Examples 29 and 30, cycloolefin was used as the main agent as the binder C. Therefore, regardless of the presence or absence of solder particles, the conduction reliability became lower after the TCT test and after the bending test. From this, it can be seen that there is an effect due to the combination of the adhesive A and the solder particles.

Also, in Comparative Examples 31 and 32, as the adhesive D, due to the polar effect of the amine-based hardener, it has adhesion to aluminum, but in the bending test, the test roller with a diameter of 10 mm or less failed to withstand conduction evaluation , And conduction reliability is low. From this, it can be seen that there is an effect due to the combination of the adhesive A and the solder particles.

On the other hand, Examples 7 to 10 were formulated with alicyclic epoxy compounds, latent Cationic hardener and acrylic resin with acrylic acid (AA) and 2-hydroxyethyl methacrylate (HEMA), so it has the characteristics of optical applications, and furthermore, it can obtain high adhesion to aluminum wiring with oxide film And excellent conduction reliability, and the melting point of the solder particles is set below the assembly temperature, so the solder particles are melted in the crimping step, the molten solder is combined with the electrode metal of the LED element, after the TCT test and after the bending test, It can also obtain higher adhesion and excellent conduction reliability.

11‧‧‧Wiring

11a‧‧‧Oxide film

12‧‧‧Sea of Epoxy Compounds

13‧‧‧ Island of acrylic resin

Claims (14)

An adhesive comprising epoxy resin, cationic catalyst, acrylic resin with a weight average molecular weight of 50,000 to 900,000, and solder particles, and the acrylic resin contains 0.5 to 10% by weight of acrylic acid and 0.5 to 10% by weight of hydroxyl group Acrylate; the above epoxy resin contains at least an alicyclic epoxy compound or a hydrogenated epoxy compound. For example, the adhesive agent according to item 1 of the patent application, wherein the amount of the solder particles is 50 parts by mass or more and less than 150 parts by mass. As for the adhesive agent according to item 1 or 2 of the patent application, the average particle diameter of the above solder particles is 3 μm or more and less than 30 μm. As for the adhesive agent according to item 1 or 2 of the patent application, the content of the acrylic resin is 1 to 10 parts by mass relative to 100 parts by mass of the epoxy resin. As for the adhesive agent of claim 3, the content of the acrylic resin is 1 to 10 parts by mass relative to 100 parts by mass of the epoxy resin. Adhesives as claimed in item 1 or 2 of the patent application, wherein the above-mentioned acrylate having a hydroxyl group is selected from 2-hydroxyethyl methacrylate, 2-hydroxypropyl methacrylate, 2-hydroxyethyl acrylate, acrylic acid One or more of the group consisting of 2-hydroxypropyl ester. Adhesive as claimed in item 3 of the patent application, wherein the acrylate having a hydroxyl group is selected from 2-hydroxyethyl methacrylate, 2-hydroxypropyl methacrylate, 2-hydroxyethyl acrylate, acrylic acid 2- One or more of the group consisting of hydroxypropyl esters. As the adhesive agent according to item 1 or 2 of the patent application, the acrylic resin includes butyl acrylate, ethyl acrylate, and acrylic nitrile. As the adhesive agent of claim 3 of the patent application, the acrylic resin includes butyl acrylate, ethyl acrylate, and nitrile acrylate. As the adhesive agent of claim 1 or 2, the above cationic catalyst is an aluminum chelate latent hardener. As the adhesive agent of claim 3, the cationic catalyst is an aluminum chelate latent hardener. If the adhesive agent of the patent application scope item 1 or 2 contains conductive particles. For example, the adhesive agent according to item 3 of the patent application contains conductive particles. A connection structure comprising: a substrate having a wiring pattern having an oxide formed on the surface; an anisotropic conductive film formed on the electrode of the wiring pattern; and an electronic component constructed on the anisotropy On the conductive film; and the anisotropic conductive film is a cured product of an anisotropic conductive adhesive, the anisotropic conductive adhesive contains an alicyclic epoxy compound or a hydrogenated epoxy compound, a cationic catalyst, and the weight average molecular weight is 50000~900000 acrylic resin, conductive particles, and solder particles, the acrylic resin contains 0.5~10wt% acrylic acid, and 0.5~10wt% acrylic acid ester with hydroxyl group.

Images